Minnesota Twins Assistant GM, Rob Antony, was kind enough to sit down with TwinsCentric’s Parker Hageman last Thursday in Fort Myers for a candid interview discussing the organization’s use of statistical analysis and preference towards traditional scouting:
TC: Defensive stats such as UZR have been made sexy by Seattle’s front office recently. Did the Twins look those kind of statistics when targeting players, like Orlando Hudson or JJ Hardy?
RA: When you look at Orlando Hudson, you can look at the UZRs and all that stuff and that’s all fine. That’s a piece. They say range is declining. Okay, what do our three scouting reports say? He still gets the ball – he’s never been fast, he’s never been a fast player – but he has all the quickness. He still has first-step quickness and he makes all the plays. So we will put more stock in that.
Just like JJ Hardy. He may not have the range he did when he was 24 at 27 now. He makes all the plays, he has enough arm and gets the ball. Our theory has always been: Make the routine plays – don’t beat yourself - and if you don’t get to a ball or two that is offset when you make all the plays. There are guys that may have 15-20 errors, a lot of them are routine errors. We’d rather have the guy that when there is a groundball to short, that’s an out. That type of thing. We scout that way, we look at it that way and we make our decisions that way.
There might be some out there that go jeez, they just traded for Hardy and signed Hudson. Our stuff says that if you look at it purely from statistical analysis and Zone Ratings and all that other stuff, defensively they’re not what they were. We’ve got people watching, saying that we know what this guy can do. That’s what we’re looking for.
TC: Do you guys have your own internal database that has this information or do you reference other websites?
We look through a lot of websites. We look at a lot of what you guys do. What you guys put together, we take all that information into consideration…the problem I have is that a few years ago when we traded [Jason] Bartlett to Tampa, they said the reason they wanted him was that he had the highest UZR, he’s got the best Zone Rating out of any shortstop. Alright. We watched the guy; we liked him and thought he was a pretty good shortstop too. A year later, people were writing “What happened to Bartlett?” His rating dropped off to 15th from number one. He wasn’t any different the next year – the stats said he was – but I’m not sure he was any different from the next year.
That’s why the defensive part of it is the hardest part for statisticians to get their hands around. And I think that’s still the case. I think that people have legitimate things that they base it on, to come up with those numbers and to rank guys out, but you treat it kind of with a grain of salt. You go with what your scouts are saying. Sometimes though you might call your scout and say “You’ve got this guy at a 6 range. The Zone Rating doesn’t back that up. You sure you want to stay with a 6 range?” He might say “I will stay with a 6 range, he’s got great first-step quickness” or he might back off and say “You know what, I didn’t see a lot of plays – he didn’t have to go to the hole that much – but he seems like he has first-step quickness.” Then he backs off from his 6 range.
It generates a good discussion. We don’t ever just sign a guy based on his written report. We call the guy that wrote the report. We want verbal confirmation; we want to hear his voice that he has conviction and belief in it. That’s a roundabout way of saying we look at all forms of information.
TC: But specifically, you are not collecting all this data, such as marking down plays that Bartlett made in the hole, or to his right?
We do not. We just hired a guy whose sole focus is statistical analysis. Gathering information and creating databases. This will be his first year. The guy that we brought in will start creating systems to build a foundation of our own that we can look at.
TC: Is that the way baseball organizations are moving in general?
I think so. This is such a competitive game and everybody is looking for that edge. We’re probably one of the last, if not the last, team to address it with a person dedicated solely to that.
TC: What took so long getting to this point of just now bringing someone on staff?
I’m not sure we bought into the stuff and we had always been so traditional. Terry Ryan was a scouting director, he was our General Manager. Mike Radcliff, Director of Player Personnel, he was a scouting director. We’ve always been really scout-oriented, people-oriented. We just have more conviction and belief in that. I think everyone has come to the realization that you cannot turn a blind-eye to that information. It is another piece of the puzzle that might give you a better informed decision.
TC: Obviously with extending Nick Blackburn and Denard Span recently, and the potential of extending Joe Mauer for eight or ten years, how do you project players for those years? Do you have a system that provides you with an expectation of what kind of results you might see in Year Four of their contract?
I’ll look at them individually. With Blackburn we looked at him and his body of work over two seasons. He averaged 200 innings, been a .500 pitcher both years and his ERA has been 4.02 or 4.05 or something like that. You look at him and then say okay, if he does that for the next three or four years, what would he make each year? Now it becomes a business decision of if you go year-to-year, what would he stand to make in arbitration next year and if he backed that up where would he go the next year? You start putting down the numbers and all the comps that he has and you base it off of if he just does what he has done. You don’t project that he is getting better – although we believe there is more in there. Instead of being an 11-11 guy, we believe he could easily be a 15-9 guy.
TC: That’s the scouting part?
That’s the scouting part. Then you take the numbers part, compare him to other players and compare what they got in arbitration. You consider what your exposure would be if you did go year-to-year with this guy. That way you limit your risk of injury if you go year-to-year. Heck, you can non-tender him if he gets injured. We looked at it and we ended up signing him to a $14 million deal. We looked at it, his comps, and what he can do and realized he would get more than that.
The other part is that we have our own checklist. That includes durability, health, makeup, does he deserve it. How is he going to handle security? There are some players that you just don’t believe that are going to keep driving to get better. They just settle in. Fortunately we haven’t had too many of those. So we’ve done a pretty good job of evaluating who the guys are.
Denard Span’s the same way. His contract is if he just does the next five years what he did his first two years. We believe there is a lot more in there. Even after that, [Span and his agent] conceded and we conceded that this is a discount. This is quite a bit less than he could have made going year-to-year if he just continued to do what he has done. In our scouting evaluations we think that he is one of the best leadoff man in the major leagues. And that was validated when some teams called and offered up substantial packages. There are other teams that look at him to same way we do. So that helps validated it a little bit because they are good teams that were willing to give up a lot for this guy. That’s the all-forms of information part of it.
TC: Prior to the Gomez trade?
We were actually pretty far down the road when a few teams called inquiring on Span.
TC: And there was a potential for a sizeable return?
Oh yeah. That makes you feel good knowing you aren’t alone out here thinking that this guy is pretty good. But there will also be people out there that think that we could have gone year-to-year and got him for that. We don’t believe so. The big one is the first year of arbitration. Because if he has a big year on a platform, he could jump up. Those guys can jump up $4 million. Now all of a sudden are we not going to be able to afford this guy a year or two down the road, are we going to have to move him because he is going to be making $10 million?
TC: So cost-control?
So you have some cost-control, you have a good player and a happy player who is thrilled we showed that confidence in him. We’re happy because we know what we’re paying him and we’re comfortable with that.
TC: How about evaluating players outside of the organization. It sounds that you grow comfortable with your own players and you can see their personalities firsthand. How do you go about finding that with players on the free market?
That’s harder and we haven’t done a lot of it. If you look at what we do, a lot of times it is just for a year or two because we still haven’t gotten away from our philosophy. When we go out and get an Orlando Hudson, we think Casilla could be an everyday second baseman. We just aren’t ready to bank on that right now. He hasn’t shown enough. If he had backed up last year like he had in 2008 then you might not do an Orlando Hudson. But for a year, now we’ll see if Casilla gets another look, another season to see if he’s ready to handle it. If not, we’ll have to go get somebody else. But there are also guys coming along. Trevor Plouffe played shortstop. If JJ Hardy shows that he is the guy and we end up keeping him for a while, Plouffe could go to second. Sometimes it acts as a stopgap.
When we went out and got Shannon Stewart everyone said “Well, there’s a rental for the rest of the year.” We ended up signing him to a three-year deal. That’s because he played and we had confidence and belief that he was our kind of guy.
We don’t do a lot of the free agents because a lot of times you wind up overpaying when you get into a competitive battle for him. It might not be what you think he’s worth. A lot of times we just back off; it’s just not worth the cost to get him. Then you have to take him for three years and he’ll be 34 years old and what’s he going to be at 35? So that’s broken down deals for us. We’ve tried to get guys at two years and an option or three and an option and they want four guaranteed. We are not going to go out that far and have a deal to become an albatross for us.
Mike Lamb didn’t work out for us when we signed him. But, you know, we weren’t in deep with a two-year deal.
TC: When you sign somebody like Lamb, who was coming out of the National League and an offensively generous ballpark in Houston, do you look at park factors?
We look at a lot of home/roads splits. We look at a lot of traditional things that I know people have gone beyond and gotten deeper. We’re big on OPS, we’re big on WHIP. We look at home/road. What’s Garret Atkins done home/road? Then you go, jeez, when there is that big of discrepancy the park obviously plays a huge role. Put him in the Metrodome and that’s equivalent to his road. How much do you want to give? What do you want to do with that? We definitely look at the factors.
With Lamb, we were looking for someone to come in and play a solid third base and, quite honestly, comparing him to what an upgrade on Nick Punto. And it was a weak free agent market; it was him and Pedro Feliz who were out there. Feliz was looking for three years at $5 million per year.
TC: I think it is fairly safe to say that I don’t think there were any projections out there that assumed Lamb would drop of as substantially as he did. [Baseball Prospectus 2008’s projection was .265/.329/.403].
TC: Working with Inside Edge, we noticed that Hardy was struggling with pitches on the outer-half of the plate, particularly from lefties, an area of strength for him in the past. Do you guys use information like that in any capacity? Do reports come in that show a hole in a swing that you might provide to [batting coach] Joe Vavra to look at further?
I think we look at some of that, but I’m not sure it plays a role in whether we acquire or not acquire. I think it’s more “Joe [Vavra], here’s all of the information we’ve got, we want to fix this guy, he’s had two good years and they’ve obviously figure out ways to pitch to him.” Joe talks to different people, finds out his swing got long, started to bar out a little bit and he’s a guy that is tough on himself when he starts to struggle. He never rebounded from that.
TC: Quite a snowball effect.
We’re hoping for a fresh start, number one. Number two, work on some of the mechanics. Joe’s looked at some footage of him. We got all that stuff, almost every at-bat from last year. They can break it down, sit with Joe and our people and go over all that stuff. That plays a role.
Just like when you get the advance scouting reports. We have an advanced guy that gets all the spray charts and all the different stuff. See how other teams have pitched him. We take our guys stuff and all the other information that we have and hand it over to Joe or [pitching coach] Rick Anderson. Here you go.
We still don’t get too far away from the strengths of our pitchers. If you start telling Nick Blackburn “Hey Blackie, these guys struggle against sliders. Pound the sliders today.” Now you’re asking him to be somebody that he’s not. In other words, if you have the good cutter going or sinker working, go ahead. We’ll take our chances with your best stuff against their stuff. And obviously you know what the weaknesses are. If you’re up two-strikes, go to the slider. You use it but you don’t change your game plan. We are aware of the weaknesses of the opposition.
And you also know where not to pitch a guy. You know, you throw it there it doesn’t matter if Lincecum, Cain or Blackburn is out there. If you throw it in that zone, he’s going to kill it. And Mauer’s great with that, he studies, he reads the charts and goes over it extensively.
TC: How about using Pitch F/X? Researchers have used the release points, various breaks and velocities to help identify both pitcher and hitter issues. Last season, after a blown save against the White Sox Mike Redmond and Ron Gardenhire both mentioned that Joe Nathan’s slider was flat. Nathan refuted this but after reviewing the Pitch F/X numbers, it reveal a decrease in the vertical drop and more horizontal movement. Do you see use for a system like this?
I think there is. And that’s another thing that the statistical analysis guy that he will need to look at. He needs to come to us and say, “You know what, this is good information. We need to utilize this.” What I do just scratches the surface. You need someone that is jumping on it. That is a prime example of that. There is more and more information out there now that doesn’t take manpower to generate, we just need to utilize it.
To read the rest of the interview, please go to Over The Baggy.